<h1 id="lisa-hunyuan-lora-l3z3klwbc">Lisa Hunyuan Lora</h1><h2 id="example-prompt:-qr7ru9jym">example prompt:</h2><ul><li><p>always starts with "<span style="color:rgb(250, 176, 5)">video of a young woman with long blonde hair and bangs"</span></p></li><li><p><span style="color:rgb(250, 176, 5)"> long dark hair</span><span style="color:#25262b"> sometimes works too</span></p></li><li><p>closed video short of a young woman with long blonde hair and bangs speaking</p></li></ul><ul><li><p>video of a young woman with long blonde hair and bangs, She is dancing and smiling to the camera. The lighting is soft and natural, highlighting her features.</p></li><li><p>video of a young woman with short black hair and bangs. she is applying lipstick in her right hand and applying it to her lips. The woman is wearing a black strapless top and her hair is styled in a bob with bangs framing her face. She has a serious expression on her face and is looking directly at the camera. The background is a solid purple color.</p></li></ul><h2 id="training-info:-atag45cu0">training info:</h2><ul><li><p>this training done on images only, somehow hunyuan support that, it will become the first frame of the video, and then somehow works like I2V, based on your training image.</p></li></ul><ul><li><p>training data is based on closed portrait, it causes the video always focus too closed to the face.</p></li></ul><ul><li><p>I feels like the model is overtrained, you can tweak down the weight when the prompt adherence is bad</p></li><li><p>training is done on RTX 4080 super with 16gb of vram and 32gb of system ram</p></li></ul><p></p><h2 id="-cu5ags2k6"></h2>

Files with SHA256: 7083319b4ef794b173a6143984b7ebb3df05e1360a0ae1b329231951bf8ece33